Robustness Battery (jfe-robustness)

When to trigger

• The result holds in the baseline but you have not tried obvious alternatives
• A referee could ask "does this survive a different measure / sample / specification?"
• You are unsure which robustness tests belong in the main text vs. the Internet Appendix
• You suspect a result might be fragile and want to find out before a referee does

Why robustness is JFE's signature

JFE's referee culture rewards thoroughness: a result is credible only after it survives a wide battery of sensible alternatives, and reviewers expect every alternative explanation to be addressed, not merely the convenient ones. A fragile result that breaks under a routine variation is a classic JFE rejection. The headline tests live in the main text; the rest go to the Internet Appendix, which JFE asks you to append to the end of the main manuscript file (see jfe-internet-appendix).

Because JFE mandates a code + non-proprietary-data deposit (Mendeley Data) at acceptance for post-2021 submissions, your robustness battery must be runnable, not just described — referees increasingly probe reproducibility after the field-wide concerns about non-replicable factors (the Fama-French factor lineage is the benchmark) and fragile staggered-DID results.

The robustness dimensions

Work through each; for each, decide main-text vs. appendix.